Binding Sites, Pockets, and Molecular Interactions

Docking does not search a whole protein at random; it places a ligand into one specific region and scores how well it fits. Before you dock anything you need to know where binding happens on the target and which physical forces hold a ligand there. This lesson builds that intuition on a real drug-target complex.

Active Versus Allosteric Sites

The active site, also called the orthosteric site, is where a protein's natural substrate binds and chemistry happens, and a competitive inhibitor works by occupying it. An allosteric pocket sits elsewhere on the surface, and a ligand bound there changes the protein's shape or dynamics without directly blocking the substrate. For docking this distinction sets your target, because a strong score in the wrong pocket is a wasted result.

Forces That Drive Binding

Binding is driven by four non-covalent forces working together. Hydrogen bonds, formed when a donor and acceptor sit about 2.7 to 3.3 angstroms apart with good geometry, give binding its directionality, while electrostatic interactions such as salt bridges between oppositely charged groups act over longer range. Hydrophobic contacts bury non-polar surface away from water and are largely entropy-driven, and short-range van der Waals interactions reward tight shape complementarity between the ligand and the pocket.

Localizing The Search

A druggable pocket is a concave, mostly enclosed cavity with enough volume to hold a small molecule, a hydrophobic lining for shape complementarity, and a few polar residues that act as hydrogen-bond anchors. You restrict docking to this region with a search box centered on the pocket, because blind docking over the whole protein is slower and produces many high-scoring poses in biologically irrelevant crevices. A clean way to define that box is to reuse the position of a ligand from a known complex.

wget -q https://files.rcsb.org/download/1HSG.pdb

# 1HSG is HIV-1 protease bound to indinavir; the drug's residue name is MK1
grep '^HETATM' 1HSG.pdb | grep 'MK1' | head -3

# the two catalytic aspartates (one per chain) that anchor the inhibitor
grep '^ATOM' 1HSG.pdb | awk '$3=="CA" && $4=="ASP" && $6==25 {print $4, $5, $6}'

# center a search box on the ligand's geometric center (PDB cols 31-54 = x,y,z)
grep '^HETATM' 1HSG.pdb | grep 'MK1' | awk '{x+=substr($0,31,8); y+=substr($0,39,8); z+=substr($0,47,8); n++} END {printf "box center  x=%.3f  y=%.3f  z=%.3f  (%d atoms)\n", x/n, y/n, z/n, n}'

HETATM 1517  N1  MK1 B 902       9.280  23.763   3.004  1.00 28.25           N
HETATM 1518  C1  MK1 B 902       9.498  23.983   4.459  1.00 30.30           C
HETATM 1519  C2  MK1 B 902      10.591  24.905   4.962  1.00 27.27           C
ASP A 25
ASP B 25
box center  x=13.073  y=22.467  z=5.557  (45 atoms)

Numbers alone do not build intuition, so look at the pocket directly. Open the same structure in a browser at rcsb.org/3d-view/1HSG, which loads the Mol* viewer, select the MK1 ligand, and display everything within about 5 angstroms of it. You will see the inhibitor sitting between the two catalytic aspartates at the dimer interface, capped by the flexible flaps that carry Ile50 from each chain, and lined by hydrophobic residues such as Val32, Ile47, Ile84, and Val82 that grip the drug by shape.